Pub Date : 2023-08-01DOI: 10.1016/j.cvdhj.2023.05.001
Arto J. Hautala PhD , Babooshka Shavazipour PhD , Bekir Afsar PhD , Mikko P. Tulppo PhD , Kaisa Miettinen PhD
Background
Health care budgets are limited, requiring the optimal use of resources. Machine learning (ML) methods may have an enormous potential for effective use of health care resources.
Objective
We assessed the applicability of selected ML tools to evaluate the contribution of known risk markers for prognosis of coronary artery disease to predict health care costs for all reasons in patients with a recent acute coronary syndrome (n = 65, aged 65 ± 9 years) for 1-year follow-up.
Methods
Risk markers were assessed at baseline, and health care costs were collected from electronic health registries. The Cross-decomposition algorithms were used to rank the considered risk markers based on their impacts on variances. Then regression analysis was performed to predict costs by entering the first top-ranking risk marker and adding the next-best markers, one by one, to build up altogether 13 predictive models.
Results
The average annual health care costs were €2601 ± €5378 per patient. The Depression Scale showed the highest predictive value (r = 0.395), accounting for 16% of the costs (P = .001). When the next 2 ranked markers (LDL cholesterol, r = 0.230; and left ventricular ejection fraction, r = -0.227, respectively) were added to the model, the predictive value was 24% for the costs (P = .001).
Conclusion
Higher depression score is the primary variable forecasting health care costs in 1-year follow-up among acute coronary syndrome patients. The ML tools may help decision-making when planning optimal utilization of treatment strategies.
{"title":"Machine learning models in predicting health care costs in patients with a recent acute coronary syndrome: A prospective pilot study","authors":"Arto J. Hautala PhD , Babooshka Shavazipour PhD , Bekir Afsar PhD , Mikko P. Tulppo PhD , Kaisa Miettinen PhD","doi":"10.1016/j.cvdhj.2023.05.001","DOIUrl":"10.1016/j.cvdhj.2023.05.001","url":null,"abstract":"<div><h3>Background</h3><p>Health care budgets are limited, requiring the optimal use of resources. Machine learning (ML) methods may have an enormous potential for effective use of health care resources.</p></div><div><h3>Objective</h3><p>We assessed the applicability of selected ML tools to evaluate the contribution of known risk markers for prognosis of coronary artery disease to predict health care costs for all reasons in patients with a recent acute coronary syndrome (n = 65, aged 65 ± 9 years) for 1-year follow-up.</p></div><div><h3>Methods</h3><p>Risk markers were assessed at baseline, and health care costs were collected from electronic health registries. The Cross-decomposition algorithms were used to rank the considered risk markers based on their impacts on variances. Then regression analysis was performed to predict costs by entering the first top-ranking risk marker and adding the next-best markers, one by one, to build up altogether 13 predictive models.</p></div><div><h3>Results</h3><p>The average annual health care costs were €2601 ± €5378 per patient. The Depression Scale showed the highest predictive value (r = 0.395), accounting for 16% of the costs (<em>P</em> = .001). When the next 2 ranked markers (LDL cholesterol, r = 0.230; and left ventricular ejection fraction, r = -0.227, respectively) were added to the model, the predictive value was 24% for the costs (<em>P</em> = .001).</p></div><div><h3>Conclusion</h3><p>Higher depression score is the primary variable forecasting health care costs in 1-year follow-up among acute coronary syndrome patients. The ML tools may help decision-making when planning optimal utilization of treatment strategies.</p></div>","PeriodicalId":72527,"journal":{"name":"Cardiovascular digital health journal","volume":"4 4","pages":"Pages 137-142"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10435951/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10105533","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.1016/j.cvdhj.2023.06.003
David Bloom MD , David Catherall MEng , Nathan Miller RN , Michael K. Southworth MSEE , Andrew C. Glatz MD, MSCE , Jonathan R. Silva PhD , Jennifer N. Avari Silva MD, FHRS
Background
CommandEP™ is a mixed reality (MXR) system for cardiac electrophysiological (EP) procedures that provides a real-time 3-dimensional digital image of cardiac geometry and catheter locations. In a previous study, physicians using the system demonstrated improved navigational accuracy. This study investigated the impact of the CommandEP system on EP procedural times compared to the standard-of-care electroanatomic mapping system (EAMS) display.
Objective
The purpose of this retrospective case-controlled analysis was to evaluate the impact of a novel MXR interface on EP procedural times compared to a case-matched cohort.
Methods
Cases from the Cardiac Augmented REality (CARE) study were matched for diagnosis and weight using a contemporary cohort. Procedural time was compared from the roll-in and full implementation cohort. During routine EP procedures, operators performed tasks during the postablation waiting phase, including creation of cardiac geometry and 5-point navigation under 2 conditions: (1) EAMS first; and (2) CommandEP.
Results
From a total of 16 CARE study patients, the 10 full implementation patients were matched to a cohort of 20 control patients (2 controls:1 CARE, matched according to pathology and age/weight). No statistical difference in total case times between CARE study patients vs control group (118 ± 29 minutes vs 97 ± 20 minutes; P = .07) or fluoroscopy times (6 ± 4 minutes vs 7 ± 6 minutes; P = .9). No significant difference in case duration for CARE study patients comparing roll-in vs full-implementation cohort (121 ± 26 minutes vs 118 ± 29 minutes; P = .96). CommandEP wear time during cases was significantly longer in full implementation cases (53 ± 24 minutes vs 24 ± 5 minutes; P = .0009). During creation of a single cardiac geometry, no significant time difference was noted between CommandEP vs EAMS (284 ± 45 seconds vs 268 ± 43 seconds; P = .1) or fluoroscopy use (9 ± 19 seconds vs 6 ± 18 seconds; P = .25). During point navigation tasks, there was no difference in total time (CommandEP 31 ± 14 seconds vs EAMS 28 ± 15 seconds; P = .16) or fluoroscopy time (CommandEP 0 second vs EAMS 0 second).
Conclusion
MXR did not prolong overall procedural time compared to a matched cohort. There was no prolongation in study task completion time. Future studies with experienced CommandEP users directly assessing procedural time and task completion time in a randomized study population would be of interest.
{"title":"Use of a mixed reality system for navigational mapping during cardiac electrophysiological testing does not prolong case duration: A subanalysis from the Cardiac Augmented REality study","authors":"David Bloom MD , David Catherall MEng , Nathan Miller RN , Michael K. Southworth MSEE , Andrew C. Glatz MD, MSCE , Jonathan R. Silva PhD , Jennifer N. Avari Silva MD, FHRS","doi":"10.1016/j.cvdhj.2023.06.003","DOIUrl":"10.1016/j.cvdhj.2023.06.003","url":null,"abstract":"<div><h3>Background</h3><p>CommandEP™ is a mixed reality (MXR) system for cardiac electrophysiological (EP) procedures that provides a real-time 3-dimensional digital image of cardiac geometry and catheter locations. In a previous study, physicians using the system demonstrated improved navigational accuracy. This study investigated the impact of the CommandEP system on EP procedural times compared to the standard-of-care electroanatomic mapping system (EAMS) display.</p></div><div><h3>Objective</h3><p>The purpose of this retrospective case-controlled analysis was to evaluate the impact of a novel MXR interface on EP procedural times compared to a case-matched cohort.</p></div><div><h3>Methods</h3><p>Cases from the Cardiac Augmented REality (CARE) study were matched for diagnosis and weight using a contemporary cohort. Procedural time was compared from the roll-in and full implementation cohort. During routine EP procedures, operators performed tasks during the postablation waiting phase, including creation of cardiac geometry and 5-point navigation under 2 conditions: (1) EAMS first; and (2) CommandEP.</p></div><div><h3>Results</h3><p>From a total of 16 CARE study patients, the 10 full implementation patients were matched to a cohort of 20 control patients (2 controls:1 CARE, matched according to pathology and age/weight). No statistical difference in total case times between CARE study patients vs control group (118 ± 29 minutes vs 97 ± 20 minutes; <em>P</em> = .07) or fluoroscopy times (6 ± 4 minutes vs 7 ± 6 minutes; <em>P</em> = .9). No significant difference in case duration for CARE study patients comparing roll-in vs full-implementation cohort (121 ± 26 minutes vs 118 ± 29 minutes; <em>P</em> = .96). CommandEP wear time during cases was significantly longer in full implementation cases (53 ± 24 minutes vs 24 ± 5 minutes; <em>P</em> = .0009). During creation of a single cardiac geometry, no significant time difference was noted between CommandEP vs EAMS (284 ± 45 seconds vs 268 ± 43 seconds; <em>P</em> = .1) or fluoroscopy use (9 ± 19 seconds vs 6 ± 18 seconds; <em>P</em> = .25). During point navigation tasks, there was no difference in total time (CommandEP 31 ± 14 seconds vs EAMS 28 ± 15 seconds; <em>P</em> = .16) or fluoroscopy time (CommandEP 0 second vs EAMS 0 second).</p></div><div><h3>Conclusion</h3><p>MXR did not prolong overall procedural time compared to a matched cohort. There was no prolongation in study task completion time. Future studies with experienced CommandEP users directly assessing procedural time and task completion time in a randomized study population would be of interest.</p></div>","PeriodicalId":72527,"journal":{"name":"Cardiovascular digital health journal","volume":"4 4","pages":"Pages 111-117"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10435945/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10105994","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.1016/j.cvdhj.2023.06.001
Constantine Tarabanis MD , Evangelos Kalampokis PhD , Mahmoud Khalil MD , Carlos L. Alviar MD , Larry A. Chinitz MD, FHRS , Lior Jankelson MD, PhD
Background
A lack of explainability in published machine learning (ML) models limits clinicians’ understanding of how predictions are made, in turn undermining uptake of the models into clinical practice.
Objective
The purpose of this study was to develop explainable ML models to predict in-hospital mortality in patients hospitalized for myocardial infarction (MI).
Methods
Adult patients hospitalized for an MI were identified in the National Inpatient Sample between January 1, 2012, and September 30, 2015. The resulting cohort comprised 457,096 patients described by 64 predictor variables relating to demographic/comorbidity characteristics and in-hospital complications. The gradient boosting algorithm eXtreme Gradient Boosting (XGBoost) was used to develop explainable models for in-hospital mortality prediction in the overall cohort and patient subgroups based on MI type and/or sex.
Results
The resulting models exhibited an area under the receiver operating characteristic curve (AUC) ranging from 0.876 to 0.942, specificity 82% to 87%, and sensitivity 75% to 87%. All models exhibited high negative predictive value ≥0.974. The SHapley Additive exPlanation (SHAP) framework was applied to explain the models. The top predictor variables of increasing and decreasing mortality were age and undergoing percutaneous coronary intervention, respectively. Other notable findings included a decreased mortality risk associated with certain patient subpopulations with hyperlipidemia and a comparatively greater risk of death among women below age 55 years.
Conclusion
The literature lacks explainable ML models predicting in-hospital mortality after an MI. In a national registry, explainable ML models performed best in ruling out in-hospital death post-MI, and their explanation illustrated their potential for guiding hypothesis generation and future study design.
{"title":"Explainable SHAP-XGBoost models for in-hospital mortality after myocardial infarction","authors":"Constantine Tarabanis MD , Evangelos Kalampokis PhD , Mahmoud Khalil MD , Carlos L. Alviar MD , Larry A. Chinitz MD, FHRS , Lior Jankelson MD, PhD","doi":"10.1016/j.cvdhj.2023.06.001","DOIUrl":"10.1016/j.cvdhj.2023.06.001","url":null,"abstract":"<div><h3>Background</h3><p>A lack of explainability in published machine learning (ML) models limits clinicians’ understanding of how predictions are made, in turn undermining uptake of the models into clinical practice.</p></div><div><h3>Objective</h3><p>The purpose of this study was to develop explainable ML models to predict in-hospital mortality in patients hospitalized for myocardial infarction (MI).</p></div><div><h3>Methods</h3><p>Adult patients hospitalized for an MI were identified in the National Inpatient Sample between January 1, 2012, and September 30, 2015. The resulting cohort comprised 457,096 patients described by 64 predictor variables relating to demographic/comorbidity characteristics and in-hospital complications. The gradient boosting algorithm eXtreme Gradient Boosting (XGBoost) was used to develop explainable models for in-hospital mortality prediction in the overall cohort and patient subgroups based on MI type and/or sex.</p></div><div><h3>Results</h3><p>The resulting models exhibited an area under the receiver operating characteristic curve (AUC) ranging from 0.876 to 0.942, specificity 82% to 87%, and sensitivity 75% to 87%. All models exhibited high negative predictive value ≥0.974. The SHapley Additive exPlanation (SHAP) framework was applied to explain the models. The top predictor variables of increasing and decreasing mortality were age and undergoing percutaneous coronary intervention, respectively. Other notable findings included a decreased mortality risk associated with certain patient subpopulations with hyperlipidemia and a comparatively greater risk of death among women below age 55 years.</p></div><div><h3>Conclusion</h3><p>The literature lacks explainable ML models predicting in-hospital mortality after an MI. In a national registry, explainable ML models performed best in ruling out in-hospital death post-MI, and their explanation illustrated their potential for guiding hypothesis generation and future study design.</p></div>","PeriodicalId":72527,"journal":{"name":"Cardiovascular digital health journal","volume":"4 4","pages":"Pages 126-132"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10435947/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10105534","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.1016/j.cvdhj.2023.04.002
Tenes J. Paul DO , Khanh-Van Tran MD, PhD , Jordy Mehawej MD, ScM , Darleen Lessard MS , Eric Ding MS , Andreas Filippaios MD , Sakeina Howard-Wilson DO , Edith Mensah Otabil BA , Kamran Noorishirazi BA , Syed Naeem MD , Alex Hamel BA , Dong Han BS , Ki H. Chon PhD , Bruce Barton PhD , Jane Saczynski PhD , David McManus MD, ScM (FHRS)
Background
The detection of atrial fibrillation (AF) in stroke survivors is critical to decreasing the risk of recurrent stroke. Smartwatches have emerged as a convenient and accurate means of AF diagnosis; however, the impact on critical patient-reported outcomes, including anxiety, engagement, and quality of life, remains ill defined.
Objectives
To examine the association between smartwatch prescription for AF detection and the patient-reported outcomes of anxiety, patient activation, and self-reported health.
Methods
We used data from the Pulsewatch trial, a 2-phase randomized controlled trial that included participants aged 50 years or older with a history of ischemic stroke. Participants were randomized to use either a proprietary smartphone-smartwatch app for 30 days of AF monitoring or no cardiac rhythm monitoring. Validated surveys were deployed before and after the 30-day study period to assess anxiety, patient activation, and self-rated physical and mental health. Logistic regression and generalized estimation equations were used to examine the association between smartwatch prescription for AF monitoring and changes in the patient-reported outcomes.
Results
A total of 110 participants (mean age 64 years, 41% female, 91% non-Hispanic White) were studied. Seventy percent of intervention participants were novice smartwatch users, as opposed to 84% of controls, and there was no significant difference in baseline rates of anxiety, activation, or self-rated health between the 2 groups. The incidence of new AF among smartwatch users was 6%. Participants who were prescribed smartwatches did not have a statistically significant change in anxiety, activation, or self-reported health as compared to those who were not prescribed smartwatches. The results held even after removing participants who received an AF alert on the watch.
Conclusion
The prescription of smartwatches to stroke survivors for AF monitoring does not adversely affect key patient-reported outcomes. Further research is needed to better inform the successful deployment of smartwatches in clinical practice.
{"title":"Anxiety, patient activation, and quality of life among stroke survivors prescribed smartwatches for atrial fibrillation monitoring","authors":"Tenes J. Paul DO , Khanh-Van Tran MD, PhD , Jordy Mehawej MD, ScM , Darleen Lessard MS , Eric Ding MS , Andreas Filippaios MD , Sakeina Howard-Wilson DO , Edith Mensah Otabil BA , Kamran Noorishirazi BA , Syed Naeem MD , Alex Hamel BA , Dong Han BS , Ki H. Chon PhD , Bruce Barton PhD , Jane Saczynski PhD , David McManus MD, ScM (FHRS)","doi":"10.1016/j.cvdhj.2023.04.002","DOIUrl":"10.1016/j.cvdhj.2023.04.002","url":null,"abstract":"<div><h3>Background</h3><p>The detection of atrial fibrillation (AF) in stroke survivors is critical to decreasing the risk of recurrent stroke. Smartwatches have emerged as a convenient and accurate means of AF diagnosis; however, the impact on critical patient-reported outcomes, including anxiety, engagement, and quality of life, remains ill defined.</p></div><div><h3>Objectives</h3><p>To examine the association between smartwatch prescription for AF detection and the patient-reported outcomes of anxiety, patient activation, and self-reported health.</p></div><div><h3>Methods</h3><p>We used data from the Pulsewatch trial, a 2-phase randomized controlled trial that included participants aged 50 years or older with a history of ischemic stroke. Participants were randomized to use either a proprietary smartphone-smartwatch app for 30 days of AF monitoring or no cardiac rhythm monitoring. Validated surveys were deployed before and after the 30-day study period to assess anxiety, patient activation, and self-rated physical and mental health. Logistic regression and generalized estimation equations were used to examine the association between smartwatch prescription for AF monitoring and changes in the patient-reported outcomes.</p></div><div><h3>Results</h3><p>A total of 110 participants (mean age 64 years, 41% female, 91% non-Hispanic White) were studied. Seventy percent of intervention participants were novice smartwatch users, as opposed to 84% of controls, and there was no significant difference in baseline rates of anxiety, activation, or self-rated health between the 2 groups. The incidence of new AF among smartwatch users was 6%. Participants who were prescribed smartwatches did not have a statistically significant change in anxiety, activation, or self-reported health as compared to those who were not prescribed smartwatches. The results held even after removing participants who received an AF alert on the watch.</p></div><div><h3>Conclusion</h3><p>The prescription of smartwatches to stroke survivors for AF monitoring does not adversely affect key patient-reported outcomes. Further research is needed to better inform the successful deployment of smartwatches in clinical practice.</p></div>","PeriodicalId":72527,"journal":{"name":"Cardiovascular digital health journal","volume":"4 4","pages":"Pages 118-125"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10435956/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10049455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-01DOI: 10.1016/j.cvdhj.2023.06.002
Joy Waughtal MPH , Thomas J. Glorioso MS , Lisa M. Sandy MA , Pamela N. Peterson MD, MSPH , Catia Chavez MPH , Sheana Bull PhD , P. Michael Ho PhD, MD , Larry A. Allen MD, MHS, FHRS
{"title":"Patient engagement with prescription refill text reminders across time and major societal events","authors":"Joy Waughtal MPH , Thomas J. Glorioso MS , Lisa M. Sandy MA , Pamela N. Peterson MD, MSPH , Catia Chavez MPH , Sheana Bull PhD , P. Michael Ho PhD, MD , Larry A. Allen MD, MHS, FHRS","doi":"10.1016/j.cvdhj.2023.06.002","DOIUrl":"10.1016/j.cvdhj.2023.06.002","url":null,"abstract":"","PeriodicalId":72527,"journal":{"name":"Cardiovascular digital health journal","volume":"4 4","pages":"Pages 133-136"},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10290239/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10105992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.1016/j.cvdhj.2023.04.004
Roberto Enrique Azcui Aparicio MD , Melinda J. Carrington PhD , Quan Huynh PhD , Jocasta Ball PhD , Thomas H. Marwick MBBS, PhD, MPH
Background
The requirement for laboratory tests to assess conventional cardiovascular disease (CVD) risk may be a barrier to the early detection and management of atherosclerosis in some population groups. A simpler risk assessment could facilitate detection of CVD.
Objectives
The association of the Fuster-BEWAT Score (FBS), Framingham Risk Score (FRS), and Pooled Cohort Equation (PCE) with the presence of carotid plaque was investigated, with the intention of developing a stepped screening process for the primary prevention of CVD.
Methods
Asymptomatic participants with a family history of premature CVD had an absolute cardiovascular disease risk (ACVDR) score calculated using the FBS, FRS, and PCE risk equations. This risk classification was compared with the presence or absence of carotid plaque on ultrasound. Prediction of carotid plaque presence by risk scores and risk factors was assessed by logistic regression and area under the curve (AUC) for discrimination and diagnostic performance. A classification and regression-tree (CART) model was obtained for stratification of risk assessment.
Results
Risk score calculation and ultrasound scanning were performed in 1031 participants, of whom 51 had carotid plaques. Participants with plaque and male sex showed higher risk (higher PCE and FRS and lower FBS, as higher scores of FBS indicate better cardiovascular health). Participants ≤50 years of age showed the FBS was a significant predictor; there was a reduced likelihood of plaque presence with a higher score (OR 0.54, 95% CI 0.39–0.75, P < .01). Higher ACVDR (evidenced by higher PCE and FRS scores and lower FBS score) was associated with an increased likelihood of carotid plaque; however, the FBS and the addition of risk factors not included in the equation showed the highest AUC (AUC = 0.76, P < .001). CART modeling showed that participants with FBS between 6 and 9 would be recommended for further risk stratification using the PCE, whereupon a PCE score ≥5% conferred an increased risk and greater possibility for plaque. Validation of the model using a different cohort showed similar risk stratification for plaque presence according to level of risk by CART analysis.
Conclusion
FBS was able to identify the presence of carotid plaque in asymptomatic individuals. Its use for initial risk delineation might improve the selection of patients for more specific and complex assessment, reducing cost and time.
{"title":"Association of cardiovascular health and risk prediction algorithms with subclinical atherosclerosis identified by carotid ultrasound","authors":"Roberto Enrique Azcui Aparicio MD , Melinda J. Carrington PhD , Quan Huynh PhD , Jocasta Ball PhD , Thomas H. Marwick MBBS, PhD, MPH","doi":"10.1016/j.cvdhj.2023.04.004","DOIUrl":"10.1016/j.cvdhj.2023.04.004","url":null,"abstract":"<div><h3>Background</h3><p>The requirement for laboratory tests to assess conventional cardiovascular disease (CVD) risk may be a barrier to the early detection and management of atherosclerosis in some population groups. A simpler risk assessment could facilitate detection of CVD.</p></div><div><h3>Objectives</h3><p>The association of the Fuster-BEWAT Score (FBS), Framingham Risk Score (FRS), and Pooled Cohort Equation (PCE) with the presence of carotid plaque was investigated, with the intention of developing a stepped screening process for the primary prevention of CVD.</p></div><div><h3>Methods</h3><p>Asymptomatic participants with a family history of premature CVD had an absolute cardiovascular disease risk (ACVDR) score calculated using the FBS, FRS, and PCE risk equations. This risk classification was compared with the presence or absence of carotid plaque on ultrasound. Prediction of carotid plaque presence by risk scores and risk factors was assessed by logistic regression and area under the curve (AUC) for discrimination and diagnostic performance. A classification and regression-tree (CART) model was obtained for stratification of risk assessment.</p></div><div><h3>Results</h3><p>Risk score calculation and ultrasound scanning were performed in 1031 participants, of whom 51 had carotid plaques. Participants with plaque and male sex showed higher risk (higher PCE and FRS and lower FBS, as higher scores of FBS indicate better cardiovascular health). Participants ≤50 years of age showed the FBS was a significant predictor; there was a reduced likelihood of plaque presence with a higher score (OR 0.54, 95% CI 0.39–0.75, <em>P</em> < .01). Higher ACVDR (evidenced by higher PCE and FRS scores and lower FBS score) was associated with an increased likelihood of carotid plaque; however, the FBS and the addition of risk factors not included in the equation showed the highest AUC (AUC = 0.76, <em>P</em> < .001). CART modeling showed that participants with FBS between 6 and 9 would be recommended for further risk stratification using the PCE, whereupon a PCE score ≥5% conferred an increased risk and greater possibility for plaque. Validation of the model using a different cohort showed similar risk stratification for plaque presence according to level of risk by CART analysis.</p></div><div><h3>Conclusion</h3><p>FBS was able to identify the presence of carotid plaque in asymptomatic individuals. Its use for initial risk delineation might improve the selection of patients for more specific and complex assessment, reducing cost and time.</p></div>","PeriodicalId":72527,"journal":{"name":"Cardiovascular digital health journal","volume":"4 3","pages":"Pages 91-100"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/fb/71/main.PMC10282005.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9767190","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.1016/j.cvdhj.2023.04.003
Alexander Schepart PharmD, MBA , Arianna Burton PharmD , Larry Durkin MBA , Allison Fuller BA , Ellyn Charap MSc , Rahul Bhambri PharmD, MBA , Faraz S. Ahmad MD, MS
Background
Numerous artificial intelligence (AI)-enabled tools for cardiovascular diseases have been published, with a high impact on public health. However, few have been adopted into, or have meaningfully affected, routine clinical care.
Objective
To evaluate current awareness, perceptions, and clinical use of AI-enabled digital health tools for patients with cardiovascular disease, and challenges to adoption.
Methods
This mixed-methods study included interviews with 12 cardiologists and 8 health information technology (IT) administrators, and a follow-on survey of 90 cardiologists and 30 IT administrators.
Results
We identified 5 major challenges: (1) limited knowledge, (2) insufficient usability, (3) cost constraints, (4) poor electronic health record interoperability, and (5) lack of trust. A minority of cardiologists were using AI tools; more were prepared to implement AI tools, but their sophistication level varied greatly.
Conclusion
Most respondents believe in the potential of AI-enabled tools to improve care quality and efficiency, but they identified several fundamental barriers to wide-scale adoption.
{"title":"Artificial intelligence–enabled tools in cardiovascular medicine: A survey of current use, perceptions, and challenges","authors":"Alexander Schepart PharmD, MBA , Arianna Burton PharmD , Larry Durkin MBA , Allison Fuller BA , Ellyn Charap MSc , Rahul Bhambri PharmD, MBA , Faraz S. Ahmad MD, MS","doi":"10.1016/j.cvdhj.2023.04.003","DOIUrl":"10.1016/j.cvdhj.2023.04.003","url":null,"abstract":"<div><h3>Background</h3><p>Numerous artificial intelligence (AI)-enabled tools for cardiovascular diseases have been published, with a high impact on public health. However, few have been adopted into, or have meaningfully affected, routine clinical care.</p></div><div><h3>Objective</h3><p>To evaluate current awareness, perceptions, and clinical use of AI-enabled digital health tools for patients with cardiovascular disease, and challenges to adoption.</p></div><div><h3>Methods</h3><p>This mixed-methods study included interviews with 12 cardiologists and 8 health information technology (IT) administrators, and a follow-on survey of 90 cardiologists and 30 IT administrators.</p></div><div><h3>Results</h3><p>We identified 5 major challenges: (1) limited knowledge, (2) insufficient usability, (3) cost constraints, (4) poor electronic health record interoperability, and (5) lack of trust. A minority of cardiologists were using AI tools; more were prepared to implement AI tools, but their sophistication level varied greatly.</p></div><div><h3>Conclusion</h3><p>Most respondents believe in the potential of AI-enabled tools to improve care quality and efficiency, but they identified several fundamental barriers to wide-scale adoption.</p></div>","PeriodicalId":72527,"journal":{"name":"Cardiovascular digital health journal","volume":"4 3","pages":"Pages 101-110"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10282011/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10070958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.1016/j.cvdhj.2023.04.001
Sana M. Al-Khatib MD, MHS, FHRS , Jagmeet P. Singh MD, DPhil, FHRS , Nassir Marrouche MD, FHRS , David D. McManus MD, PhD, ScM, FHRS , Andrew D. Krahn MD, FHRS , Patricia Blake FASAE, CAE
{"title":"The inaugural 2022 HRX meeting: A patient-centered digital health meeting for the acceleration of cardiovascular innovation","authors":"Sana M. Al-Khatib MD, MHS, FHRS , Jagmeet P. Singh MD, DPhil, FHRS , Nassir Marrouche MD, FHRS , David D. McManus MD, PhD, ScM, FHRS , Andrew D. Krahn MD, FHRS , Patricia Blake FASAE, CAE","doi":"10.1016/j.cvdhj.2023.04.001","DOIUrl":"10.1016/j.cvdhj.2023.04.001","url":null,"abstract":"","PeriodicalId":72527,"journal":{"name":"Cardiovascular digital health journal","volume":"4 3","pages":"Pages 69-71"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/01/a4/main.PMC10282004.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10059539","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The use of 12-lead electrocardiogram (ECG) is common in routine primary care, however it can be difficult for less experienced ECG readers to adequately interpret the ECG.
Objective
To validate a smartphone application (PMcardio) as a stand-alone interpretation tool for 12-lead ECG in primary care.
Methods
We recruited consecutive patients who underwent 12-lead ECG as part of routinely indicated primary care in the Netherlands. All ECGs were assessed by the PMcardio app, which analyzes a photographed image of 12-lead ECG for automated interpretation, installed on an Android platform (Samsung Galaxy M31) and an iOS platform (iPhone SE2020). We validated the PMcardio app for detecting any major ECG abnormality (MEA, primary outcome), defined as atrial fibrillation/flutter (AF), markers of (past) myocardial ischemia, or clinically relevant impulse and/or conduction abnormalities; or AF (key secondary outcome) with a blinded expert panel as reference standard.
Results
We included 290 patients from 11 Dutch general practices with median age 67 (interquartile range 55–74) years; 48% were female. On reference ECG, 71 patients (25%) had MEA and 35 (12%) had AF. Sensitivity and specificity of PMcardio for MEA were 86% (95% CI: 76%–93%) and 92% (95% CI: 87%–95%), respectively. For AF, sensitivity and specificity were 97% (95% CI: 85%–100%) and 99% (95% CI: 97%–100%), respectively. Performance was comparable between Android and iOS platform (kappa = 0.95, 95% CI: 0.91–0.99 and kappa = 1.00, 95% CI: 1.00–1.00 for MEA and AF, respectively).
Conclusion
A smartphone app developed to interpret 12-lead ECGs was found to have good diagnostic accuracy in a primary care setting for major ECG abnormalities, and near-perfect properties for diagnosing AF.
{"title":"Diagnostic accuracy of the PMcardio smartphone application for artificial intelligence–based interpretation of electrocardiograms in primary care (AMSTELHEART-1)","authors":"Jelle C.L. Himmelreich MD, MSc , Ralf E. Harskamp MD, PhD","doi":"10.1016/j.cvdhj.2023.03.002","DOIUrl":"10.1016/j.cvdhj.2023.03.002","url":null,"abstract":"<div><h3>Background</h3><p>The use of 12-lead electrocardiogram (ECG) is common in routine primary care, however it can be difficult for less experienced ECG readers to adequately interpret the ECG.</p></div><div><h3>Objective</h3><p>To validate a smartphone application (PMcardio) as a stand-alone interpretation tool for 12-lead ECG in primary care.</p></div><div><h3>Methods</h3><p>We recruited consecutive patients who underwent 12-lead ECG as part of routinely indicated primary care in the Netherlands. All ECGs were assessed by the PMcardio app, which analyzes a photographed image of 12-lead ECG for automated interpretation, installed on an Android platform (Samsung Galaxy M31) and an iOS platform (iPhone SE2020). We validated the PMcardio app for detecting any major ECG abnormality (MEA, primary outcome), defined as atrial fibrillation/flutter (AF), markers of (past) myocardial ischemia, or clinically relevant impulse and/or conduction abnormalities; or AF (key secondary outcome) with a blinded expert panel as reference standard.</p></div><div><h3>Results</h3><p>We included 290 patients from 11 Dutch general practices with median age 67 (interquartile range 55–74) years; 48% were female. On reference ECG, 71 patients (25%) had MEA and 35 (12%) had AF. Sensitivity and specificity of PMcardio for MEA were 86% (95% CI: 76%–93%) and 92% (95% CI: 87%–95%), respectively. For AF, sensitivity and specificity were 97% (95% CI: 85%–100%) and 99% (95% CI: 97%–100%), respectively. Performance was comparable between Android and iOS platform (kappa = 0.95, 95% CI: 0.91–0.99 and kappa = 1.00, 95% CI: 1.00–1.00 for MEA and AF, respectively).</p></div><div><h3>Conclusion</h3><p>A smartphone app developed to interpret 12-lead ECGs was found to have good diagnostic accuracy in a primary care setting for major ECG abnormalities, and near-perfect properties for diagnosing AF.</p></div>","PeriodicalId":72527,"journal":{"name":"Cardiovascular digital health journal","volume":"4 3","pages":"Pages 80-90"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10282008/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10070951","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.1016/j.cvdhj.2023.03.003
Edmond M. Cronin MB, BCh, BAO, FHRS , Joseph C. Green BS , Jeff Lande PhD , Thomas R. Holmes PhD , Daniel Lexcen PhD , Tyler Taigen MD, FHRS
Background
Remote monitoring of cardiac implantable electric devices improves patient outcomes and experiences. Alert-based systems notify physicians of clinical or device issues in near real-time, but their effectiveness is contingent upon device connectivity.
Objective
To assess patient connectivity by analyzing alert transmission times from patient transceivers to the CareLink network.
Methods
Alert transmissions were retrospectively gathered from a query of the United States de-identified Medtronic CareLink database. Alert transmission time was defined as the duration from alert occurrence to arrival at the CareLink network and was analyzed by device type, alert event, and alert type. Using data from previous studies, we computed the benefit of daily connectivity checks.
Results
The mean alert transmission time was 14.8 hours (median = 6 hours), with 90.9% of alert transmissions received within 24 hours. Implantable pulse generators (17.0 ± 40.2 hours) and cardiac resynchronization therapy-pacemakers (17.2 ± 42.5 hours) had longer alert transmission times than implantable cardioverter-defibrillators (13.7 ± 29.5 hours) and cardiac resynchronization therapy-defibrillators (13.5 ± 30.2 hours), but the median time was 6 hours for all 4 device types. There were differences in alert times between specific alert events. Based on our data and previous studies, daily connectivity checks could improve daily alert transmission success by 8.5% but would require up to nearly 800 additional hours of staff time on any given day.
Conclusion
Alert transmission performance from Medtronic devices was satisfactory, with some delays likely underscored by patient connectivity issues. Daily connectivity checks could provide some improvement in transmission success at the expense of increased clinic burden.
{"title":"Performance of alert transmissions from cardiac implantable electronic devices to the CareLink network: A retrospective analysis","authors":"Edmond M. Cronin MB, BCh, BAO, FHRS , Joseph C. Green BS , Jeff Lande PhD , Thomas R. Holmes PhD , Daniel Lexcen PhD , Tyler Taigen MD, FHRS","doi":"10.1016/j.cvdhj.2023.03.003","DOIUrl":"10.1016/j.cvdhj.2023.03.003","url":null,"abstract":"<div><h3>Background</h3><p>Remote monitoring of cardiac implantable electric devices improves patient outcomes and experiences. Alert-based systems notify physicians of clinical or device issues in near real-time, but their effectiveness is contingent upon device connectivity.</p></div><div><h3>Objective</h3><p>To assess patient connectivity by analyzing alert transmission times from patient transceivers to the CareLink network.</p></div><div><h3>Methods</h3><p>Alert transmissions were retrospectively gathered from a query of the United States de-identified Medtronic CareLink database. Alert transmission time was defined as the duration from alert occurrence to arrival at the CareLink network and was analyzed by device type, alert event, and alert type. Using data from previous studies, we computed the benefit of daily connectivity checks.</p></div><div><h3>Results</h3><p>The mean alert transmission time was 14.8 hours (median = 6 hours), with 90.9% of alert transmissions received within 24 hours. Implantable pulse generators (17.0 ± 40.2 hours) and cardiac resynchronization therapy-pacemakers (17.2 ± 42.5 hours) had longer alert transmission times than implantable cardioverter-defibrillators (13.7 ± 29.5 hours) and cardiac resynchronization therapy-defibrillators (13.5 ± 30.2 hours), but the median time was 6 hours for all 4 device types. There were differences in alert times between specific alert events. Based on our data and previous studies, daily connectivity checks could improve daily alert transmission success by 8.5% but would require up to nearly 800 additional hours of staff time on any given day.</p></div><div><h3>Conclusion</h3><p>Alert transmission performance from Medtronic devices was satisfactory, with some delays likely underscored by patient connectivity issues. Daily connectivity checks could provide some improvement in transmission success at the expense of increased clinic burden.</p></div>","PeriodicalId":72527,"journal":{"name":"Cardiovascular digital health journal","volume":"4 3","pages":"Pages 72-79"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/fe/36/main.PMC10282010.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10070953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号